APEX-Agents · Investment Banking
world227_tg_06
APEX-Agents task world227_tg_06 in AI Agents for Infrastructure Finance. Compare dual-harness agent runs across models, scores, and public traces.
Task prompt
What the agent was asked to do
Use the MFC model we made. MFC has chosen to proceed with a debt-only refinancing at market rates (Option A / Case 1). At the start of each year, it decides to issue new debt up to a maximum of 4.0x gross leverage in total debt capacity (LTM EBITDA), with the proceeds used to fund special dividends to its shareholders. Please calculate the amount of special dividends received until the end of the projection period in FY32. Provide your response back to me here, rounded to the millions of dollars.
Published trajectories
Agent runs on this task
Curated dual-harness runs (parsed + original sandbox). Best scored run per model.
| Model | Harness | Score | Result | Links |
|---|---|---|---|---|
| GPT-5.4 mini | dual | 1/1 | Pass | Run detailsPublic trace |
| GPT-5.4 nano | dual | 1/1 | Pass | Run detailsPublic trace |
| Gemini 3.1 Pro | dual | 0/1 | Fail | Run detailsPublic trace |
| GPT-5.5 | dual | 0/1 | Fail | Run detailsPublic trace |
Grading rubric
Rubric criteria
Runs are graded against these criteria. Open a run for model-specific verdicts.
States the total amount of special dividends received until the end of the projection period in FY32 is $1,783 million